Checker
PassAudited by VirusTotal on May 11, 2026.
Overview
Type: OpenClaw Skill Name: checker Version: 1.0.0 The skill bundle defines a 'checker' agent for QA and peer review. The `SKILL.md` provides clear, structured instructions for the agent's role, workflow, and output format. It explicitly limits the agent's actions (e.g., 'Don't rewrite content yourself'). File writing is restricted to a designated workspace directory (`/home/ubuntu/.openclaw/workspace/artifacts/checker/`), which is standard for agent output. There are no instructions for data exfiltration, malicious execution, persistence, or prompt injection attempts against the agent itself. The content aligns entirely with its stated purpose without any high-risk behaviors.
Findings (0)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Reviewed material and QA comments may remain saved locally after the review is complete.
The skill stores review reports in a persistent workspace folder. This is expected for a QA artifact, but those reports may include details about content, research, outreach, or image reviews.
Outputs: review verdicts saved to artifacts/checker/ ... All reviews: `/home/ubuntu/.openclaw/workspace/artifacts/checker/`
Avoid submitting highly sensitive drafts unless you are comfortable with the review being saved in the disclosed artifacts folder, or clear that folder when appropriate.
Drafts or review results may be shared within the agent workflow rather than staying only in the immediate conversation.
The skill is designed to review outputs from other agents and report a verdict to another named agent. This is consistent with the stated QA workflow, but it means content may move between agents in the workflow.
Receive output from another agent ... Report verdict to Cello
Use it for intended pre-delivery QA and ensure the reviewed content is appropriate to share with the named workflow agents.